Re: [HACKERS] increasing the default WAL segment size

Поиск
Список
Период
Сортировка
От David Steele
Тема Re: [HACKERS] increasing the default WAL segment size
Дата
Msg-id c333b1e2-2043-af26-5eba-bd9d2798adaa@pgmasters.net
обсуждение исходный текст
Ответ на Re: [HACKERS] increasing the default WAL segment size  (Beena Emerson <memissemerson@gmail.com>)
Список pgsql-hackers
On 4/7/17 2:59 AM, Beena Emerson wrote:
> I ran tests and following are the details:
> 
> Machine details:
> Architecture:          ppc64le
> Byte Order:            Little Endian
> CPU(s):                192
> On-line CPU(s) list:   0-191
> Thread(s) per core:    8
> Core(s) per socket:    1
> Socket(s):             24
> NUMA node(s):          4
> Model:                 IBM,8286-42A
> 
> clients>          16                      32                       64  
>                  128
> size
> 16MB      18895.63486     28799.48759     37855.39521     27968.88309
> 32MB      18313.1461      29201.44954     40733.80051     32458.74147
> 64 MB    18055.73141     30875.28687     42713.54447     38009.60542
> 128MB   18234.31424     33208.65419     48604.5593      45498.27689
> 256MB    19524.36498     35740.19032     54686.16898     54060.11168
> 512MB     20351.90719     37426.72174     55045.60719     56194.99349
> 1024MB   19667.67062     35696.19194     53666.60373     54353.0614
> 
> I did not get any degradation, in fact, higher values showed performance
> improvement for higher client count.

This submission has been moved to CF 2017-07.

-- 
-David
david@pgmasters.net



В списке pgsql-hackers по дате отправления:

Предыдущее
От: Andres Freund
Дата:
Сообщение: Re: [HACKERS] Running make check-world in buildfarm (was Re:[COMMITTERS] pgsql: Use SASLprep to normalize passwords for SCRAMauthentication.)
Следующее
От: David Steele
Дата:
Сообщение: Re: [HACKERS] Vacuum: allow usage of more than 1GB of work mem